Re: IPA speech synthesizer
| From: | Philip Newton <philip.newton@...> |
|---|
| Date: | Friday, February 20, 2009, 19:26 |
|---|
2009/2/20 Roger Mills <romiltz@...>:
> Would it not be poassible to create a machine that could read and reproduce
> spectrograms? (Or is that what Alex's "bigrams" meant?)
I interpreted "bigrams" as combinations of phones.
So that, for example, "conlang" would be, conceptually, produced by
splicing together the bigrams for [kQ] [Qn] [nl] [l&] [&N]. (Not sure
whether [#k] and [N#] would also get bigrams; possibly so.)
This would go some way toward taking care of the "troughs" that BPJ
referred to, since you don't have, say, [&] spoken in isolation but as
part of the bigrams [l&] and [&N], which take care of the transitions.
(To produce arbitrary texts, you'd obviously need thousands of bigrams
- many more than for any individual language where the list would be
constrained by the phonotactics.)
Cheers,
--
Philip Newton <philip.newton@...>
Replies